Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

HIP-119: Location Trust Score from maximum asserted distance difference #840

Merged
merged 19 commits into from
Jul 17, 2024

Conversation

michaeldjeffrey
Copy link
Contributor

@michaeldjeffrey michaeldjeffrey commented Jul 10, 2024

HIP-119 introduces a new way for a radio to be ineligible for receiving Service Provider boosted rewards.

When validating a Heartbeat, a new set of distances for indoor and outdoor radios are used to determine a Location Trust Multiplier.

BoostedHexStatus has a new Status!!

BoostedHexStatus::AverageAssertedDistanceOverLimit(Decimal) means a Radio is not eligible to receive service provider boosted rewards because the average distance heartbeat locations from the radio's asserted location is greater than 50 meters.

Average vs. Maximum

"Maximum Asserted Distance for Boosted Hexes" is the heading used in HIP-119. A followup question was asked in the #hip-119-closing-gaming-loopsholes-within-the-mobile-network Discord channel.

"[...] is that the average of distance_to_asserted over the day or the max distance_to_asserted?"

And was answered by the HIP author.

"I’ll leave that up to you guys, whichever you think is the better approach"

This implementation has chosen to go with the average distance from an asserted location being greater than 50 meters, rather than any individual distance from the asserted location being greater than 50 meters.

If a radio spends more time than not within 50 meters of its asserted location, the benefit of the doubt would say it is most likely in that location. A temporary excursion outside of 50 meters through the means of SkyHook or other location reporting services will not be penalized as harshly as possible.

Service Provider Hex Boosting and Location Trust

It was previously the case that a radio covering a boosted hex would lower the location trust score of any LocationTrust that was more then 50 meters from the asserted location.

Now, regardless of covering boosted hexes, LocationTrust will not be changed.

A Radio will not receive boosted rewards when greater than 50 meters away from it's asserted location on average. A Wifi Radio will not receive boosted rewards if it's naturally occuring location_trust_multiplier falls below 0.75x as stated in hip-93.

Items and answers

  • valid heartbeats, should they completely lower a trust score? Yes
  • furthest covered hex, should a single hex exceeding the max_distance_to_coverage setting completely lower a trust score? Yes
  • Do you think I could pull off this jacket Resounding Yes

HIP-119 introduces a new set of tables for location trust scores based on `(radio_type, distance_to_asserted)`.

This also allows for a new minimum trust score multiplier of `0x`.
This value is now contained within the function `asserted_distance_to_trust_multiplier` in the coverage point calculator.
All location trust scores used to be 1.0 or 0.25. HIP-119 adds 0.00 as a multiplier possibility based on distance.
Location Trust Scores are no longer reduced based on the presence of a
boosted hex. However, having an average distance from an asserted
location past 50m can cause a radio to be ineligible for boosted rewards.
Both constants have to do with service provider boosting in regards to a
radios location trust scores. Namespacing allows for not needing to
shove all possible context into a top level name.
It didn't feel quite correct to have half the service provider boosting
code in lib.rs and the other in location.rs.

The constants have to with location trust scores, but they do not get
used there.
The distance to asserted no longer changes trust scores in the presence
of a boosted hex. Service provider boosting eligibility is determined
from the distance to asserted, that is tested at the top level of this crate.
The only difference was the location distance from asserted and assigned
location trust multipliers.

Refactoring heartbeat seeding further is left as an exercise to the next
person who has a need to change these tests.
…osting

HIP-119 removes the part of calculating coverage points that degrades a
trust score when a radio covers a boosted hex and is more than 50m away
from their asserted location.

Now, being +50m away from an asserted location makes a radio not
eligible for receiving service provider boosted rewards. They continue
to receive the full force of their location trust score.

Being an Indoor radio 100m away from an asserted location is enough to
keep a location trust multiplier of 1.0x, but not receive boosted rewards.
Adding links to multiple places so you don't have to know a secret
location where they exist.
@bbalser
Copy link
Contributor

bbalser commented Jul 11, 2024

The answer to all 3 questions is yes. If no location_validate_timestamp is present or the max_distance to coverage_object is > 3000m than the location_trust_score_multiplier should be 0.

And that jacket would look great!

coverage_point_calculator/src/service_provider_boosting.rs Outdated Show resolved Hide resolved
mobile_verifier/src/heartbeats/mod.rs Outdated Show resolved Hide resolved
coverage_point_calculator/src/location.rs Outdated Show resolved Hide resolved
Both were answered yes
CBRS is always trusted for location. The location trust score is intercepted earlier when validating heartbeats. If for some reason CBRS radios do make it to this function, they will receive a good trust score anyways.
@michaeldjeffrey michaeldjeffrey merged commit 3e142e9 into main Jul 17, 2024
17 checks passed
@michaeldjeffrey michaeldjeffrey deleted the mj/hip-119-location-trust branch July 17, 2024 16:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants